Open joleyline opened 4 years ago
Is there a way to raise the temp safety limit from 70c to 80 or 90c to mitigate this effect? Ryzen chips should be safe operating even at 95-100c.
Unfortunately 69c seems to be the maximum that's allowed by the asus driver for the fanless mode (I don't think we can increase that). Btw: I don't see those spikes on my computer (at least with CPU Boost mode set to Disabled).
Also would it be possible to have the custom profiles support linked fan speeds like in the windows default plan? The heatpipes are shared between the CPU and GPU and the two fans running at low-medium speeds are much better acoustically than just forcing the CPU fan to run at full power.
(I'm not sure whether I understand that question). The GPU fan should be running too above a certain (configurable) temperature (to be honest I never tested whether it really does). ... and as you said, the CPU and GPU heatpipes are connected, so when the CPU gets too hot, the GPU gets hot too and thus the GPU fan should be running too.
Unfortunately 69c seems to be the maximum that's allowed by the asus driver for the fanless mode
Are you sure? You correct the values lower than that in your program, so I can't check, but I bet sometimes for some unknown reason my laptop stays on 0 rpm even on 80c. But other than those time it kills me with whining though :\
and as you said, the CPU and GPU heatpipes are connected, so when the CPU gets too hot, the GPU gets hot too and thus the GPU fan should be running too
Sadly it works another way - while only one fan is running (which is usually the case) cooling becomes much worse. Syncing their speeds is a much better way, is this possible though? Standard profiles work this way
This may not be a problem for those who disable boost.
So with boost enabled, zen chips often ramp up to full voltage for single cores - which causes a very short spike in temperature. These are often very temporary spikes that don't show up on most temperature monitoring tools. The problem is that the actual temp sensors that inform the fans will pick up on these instantaneous readings and ramp up the fans only to immediately stop. So you have these insanely quick spikes from ~40-50c to ~70c that drop just as quickly afterwards. There really is no use for the fan to react to them.
Is there a way to raise the temp safety limit from 70c to 80 or 90c to mitigate this effect? Ryzen chips should be safe operating even at 95-100c.
Also would it be possible to have the custom profiles support linked fan speeds like in the windows default plan? The heatpipes are shared between the CPU and GPU and the two fans running at low-medium speeds are much better acoustically than just forcing the CPU fan to run at full power.